Omnilingual MT: Machine Translation for 1,600 Languages
We present OMT, the first MT system supporting more than 1,600 languages, where 1B–8B parameter specialized models match or exceed a 70B LLM baseline, with strong generalization to …
BOUQuET: dataset, Benchmark and Open initiative for Universal Quality Evaluation in Translation
We introduce BOUQuET, a multi-way, multicentric, multi-domain, paragraph-level dataset and benchmark for machine translation evaluation, designed for crowd-sourced expansion to …
Speech Is More than Words: Do Speech-to-Text Translation Systems Leverage Prosody?
We investigate whether speech-to-text translation systems utilize prosody by introducing a new benchmark, ContraProSt. Our findings show that while models represent prosody …